Weighting Procedures for the 2002 NSAF

Report No. 3 in the 2002 NSAF Methodology Series

The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.

Note: This report is available in its entirety in the Portable Document Format (PDF).

1. INTRODUCTION

This report describes the weighting procedures used in Round 3 of the National Survey of America's Families (NSAF), conducted in 2002. Those familiar with the methodology reports for the 1997 and 1999 NSAFs will note similarities between the present and earlier reports. This report aims to explain the weighting in heuristic steps that will make technical matter accessible. As one of the goals of NSAF is to produce estimates of change, Round 3 estimation is compared to estimation procedures in Rounds 1 and 2.

Chapter 1 of this report is written for the general reader and describes the weighting and estimation process in broad stokes. Section 1.1 gives a general overview of the survey. Section 1.2 describes the goals of weighting and some specifics about how it is carried out. Section 1.3 discusses how the weights are best used and gets into some detail on estimates of change. Section 1.4 talks about the contents of other chapters in this report. Other technical topics, for example, variance estimation, data editing, and imputation, are discussed in other reports in this methodology series.

1.1 Overview of the Survey

The NSAF collected information on the economic, health, and social dimensions of the wellbeing of children, adults under the age of 65, and their families in 13 states and the balance of the Nation (see figure 1-1). This section briefly outlines the major sample design features. More details on the design are given in the 2002 NSAF Sample Design, Report No. 2, in this series.

In Round 1 and Round 2, Milwaukee County in Wisconsin was a separate study area that had its own sample. A separate sample from the balance of Wisconsin was selected in these rounds to produce estimates for the entire state of Wisconsin by combining Milwaukee and the balance of Wisconsin. In Round 3, the separate Milwaukee County study area was eliminated and the entire state of Wisconsin was treated as a single study area as shown in figure 1-1.

A key goal of the survey was to obtain social and economic information about children in low-income households (income below 200 percent of the poverty threshold). Similar data on children in all households, low-income adults under age 65, and on other adults under age 65 were also required. After Round 1, an additional goal was to precisely estimate changes between the rounds.

The sampling design consisted of two components: (1) a random digit dial (RDD) component for households with telephones and (2) an area sample conducted in person for those households without telephones. This dual-frame approach is described in detail in Waksberg, et al. (1997). The Round 3 area sample is very different from that used in the previous rounds. In Round 1 and Round 2, separate area samples were selected in each study area and for the balance of the Nation. The area and RDD samples were combined to produce estimates of both telephone and nontelephone households for each study area and nationally. In Round 3, the area sample was revised so that only a national sample was selected rather than area sample from each study area. Consequently, the area sample sizes were reduced and a new weighting scheme was introduced in Round 3. The reduction in the size of the area sample significantly reduced the cost of data collection.

As a result of the changes in the sample design for Round 3, the procedure of combining the estimates from the RDD sample and area sample for each study area used in the previous rounds was no longer applicable. Essentially, estimates from the national sample that combines the RDD and the area sample could be produced using methods similar to those used in previous rounds. However, the study area estimates required new estimation procedures and corresponding weights because no area samples were available for the study areas. Thus, one set of weights was created for the national sample and a separate set of weights was created for the study areas. The methodology for each set of weights is described later in this report.

Many of the sampling procedures in Round 3 that affect the weighting procedures were used in previous rounds. Low-income households and households with children were sampled at higher rates than other households. Income status and family status were determined in a screening interview and this information was then used to subsample households. In the area sample, blocks with very high telephone coverage rates as of the 1990 Census were excluded to reduce costs, and households containing only adults age 65 and over were also excluded. Household members were subsampled in an effort to reduce respondent burden. In households with more than one child under age 6, only one child in this age range was randomly selected. Independently, only one child was also selected for households that had more than one child in the ages 6 to 17.

As in previous rounds, child interviews are conducted with the most knowledgeable adult (MKA), that is the adult in the household who is most knowledgeable about the health care, education, and well-being of the sampled child. If two children were sampled, it is possible to have different MKAs for each child, although it was more common to have a single respondent (usually the mother) for both children. The MKA responded for the sampled child, for him/herself, and for his/her partner/spouse. Most questions asked about the MKA were repeated in reference to the spouse/partner; however, some questions on health insurance and health care utilization were asked in reference to only one of the two. The target of these questions was randomly assigned to either the MKA or his/her spouse/partner. There were also some questions related to feelings, religious activities, and opinions that were asked only about the MKA.

Other adults in households with children were subsampled, as were adults in adult-only households. Adults were eligible only if they would not have been MKAs for other children in the household if those children had been selected. Self-response was required for sampled adults, and data were also collected about his/her spouse/partner (if living in the same household). Data were not collected directly from the spouse of a sampled adult. As in the MKA interview, some questions related to feelings, religious activities, and opinions were asked only about the sampled adult.

Note: This report is available in its entirety in the Portable Document Format (PDF).

The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.

Usage, posting and reprint of materials on the UI web site:

Most publications may be downloaded free of charge from the web site in PDF format. This information may be used and copies made for research, academic, policy or other non-commercial purposes. Proper attribution is required.

Copyright of the written materials contained within the Urban Institute website is owned or controlled by the Urban Institute. Posting UI research papers on other websites is permitted subject to prior approval from the Urban Institute—contact paffairs@urban.org.

If you are unable to access or print the PDF document please contact us or call the Publications Office at (202) 261-5687.